Comment by zapnuk

US models align with our "average" (western) values. If we outsource thinking by using LLMs, why would we outsource it to an LLM that doesn't have our values encoded in it?